During the CMIP6 process, ES-DOC in co-ordination with MIP chairs & modelling groups, will create a wide array of documentation. Document creation is time-dependent, e.g. model documentation must occur after he models have been configured. It is useful therefore to consider a timeline that breaks down the documentation process into several phases. We hereby break down the process into 4 phases – however note that some of the phases run contiguously.
| Phase | Timeline | Documentation | ES-DOC PI(s) |
|---|---|---|---|
| 1 | Operational (since March 2018) | Experiments | Charlotte Pascoe @ STFC, UK |
| 2 | Beta testing completed (Nov 2018) | Models | Eric Guilyardi @ IPSL, FR |
| 3 | Q1 2019 onwards | Ensemble / Conformance | David Hassell @ NCAS, UK |
| 4 | Q2 2018 onwards | Performance | David Hassell @ NCAS, UK |
| 5 | Complete | Errata | Guillaume Levavasseur @ IPSL, FR |
Phase 1: Experiments
During this initial phase documentation of the CMIP6 protocols will be the primary focus. This work is co-ordinated by ES-DOC in association with WCRP and the various MIP chairs. Gathering detailed information regarding each experiment and the set of participating MIPs is the core of this work. Particularly difficult is defining experimental requirements, e.g. forcing constraints, and mapping these requirements to experiments and thus MIPs. Further information is found here.
Phase 2: Models
Once groups decide upon the set of model configurations they will use to run simulations, then they can be invited to document these configurations. In co-ordination with realm experts, ES-DOC has defined so-called specialisations, i.e. questions upon scientific themes. The groups will have at their disposal an array of tools with which they can document their model configurations. Further information is found here.
Phase 3: Ensembles & Simulations
As the simulations are running and CMIP6 datasets begin to be published it will be necessary to document ensembles. In particular it is scientifically useful to understand how the groups have conformed to the set of experimental requirements. This helps to understand ensemble variance and perform forcing comparisons. ES-DOC is working alongside ESG-F to integrate ensemble documentation creation alongside dataset publication. Groups will subsequently use ES-DOC supplied spreadsheets to precisely define how there ensembles conformed (or not) to the experimental requirements defined in the protocols. Further information is found here.
Phase 4: Post Simulation
Inevitably post-simulation brings the understanding that some published datasets may have issues. Documenting these issues during CMIP6 will be both simplified and standardised via the ES-DOC Dataset Errata system. This includes client tools for creating errata reports, plus online search & view tools for quickly identifying errata by dataset, model, experiment, variable …etc. Furthermore researchers interested in quantifying computational performance will be able to document the performance of the machines upon which the simulations were run.